@inproceedings{yakovlev-etal-2023-gec,
title = "{GEC}-{D}e{P}en{D}: Non-Autoregressive Grammatical Error Correction with Decoupled Permutation and Decoding",
author = "Yakovlev, Konstantin and
Podolskiy, Alexander and
Bout, Andrey and
Nikolenko, Sergey and
Piontkovskaya, Irina",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.86",
doi = "10.18653/v1/2023.acl-long.86",
pages = "1546--1558",
abstract = "Grammatical error correction (GEC) is an important NLP task that is currently usually solved with autoregressive sequence-to-sequence models. However, approaches of this class are inherently slow due to one-by-one token generation, so non-autoregressive alternatives are needed. In this work, we propose a novel non-autoregressive approach to GEC that decouples the architecture into a permutation network that outputs a self-attention weight matrix that can be used in beam search to find the best permutation of input tokens (with auxiliary {\textless}ins{\textgreater} tokens) and a decoder network based on a step-unrolled denoising autoencoder that fills in specific tokens. This allows us to find the token permutation after only one forward pass of the permutation network, avoiding autoregressive constructions. We show that the resulting network improves over previously known non-autoregressive methods for GEC and reaches the level of autoregressive methods that do not use language-specific synthetic data generation methods. Our results are supported by a comprehensive experimental validation on the ConLL-2014 and BEA datasets and an extensive ablation study that supports our architectural and algorithmic choices.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yakovlev-etal-2023-gec">
<titleInfo>
<title>GEC-DePenD: Non-Autoregressive Grammatical Error Correction with Decoupled Permutation and Decoding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Konstantin</namePart>
<namePart type="family">Yakovlev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Podolskiy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrey</namePart>
<namePart type="family">Bout</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sergey</namePart>
<namePart type="family">Nikolenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Irina</namePart>
<namePart type="family">Piontkovskaya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Grammatical error correction (GEC) is an important NLP task that is currently usually solved with autoregressive sequence-to-sequence models. However, approaches of this class are inherently slow due to one-by-one token generation, so non-autoregressive alternatives are needed. In this work, we propose a novel non-autoregressive approach to GEC that decouples the architecture into a permutation network that outputs a self-attention weight matrix that can be used in beam search to find the best permutation of input tokens (with auxiliary \textlessins\textgreater tokens) and a decoder network based on a step-unrolled denoising autoencoder that fills in specific tokens. This allows us to find the token permutation after only one forward pass of the permutation network, avoiding autoregressive constructions. We show that the resulting network improves over previously known non-autoregressive methods for GEC and reaches the level of autoregressive methods that do not use language-specific synthetic data generation methods. Our results are supported by a comprehensive experimental validation on the ConLL-2014 and BEA datasets and an extensive ablation study that supports our architectural and algorithmic choices.</abstract>
<identifier type="citekey">yakovlev-etal-2023-gec</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.86</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.86</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>1546</start>
<end>1558</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GEC-DePenD: Non-Autoregressive Grammatical Error Correction with Decoupled Permutation and Decoding
%A Yakovlev, Konstantin
%A Podolskiy, Alexander
%A Bout, Andrey
%A Nikolenko, Sergey
%A Piontkovskaya, Irina
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F yakovlev-etal-2023-gec
%X Grammatical error correction (GEC) is an important NLP task that is currently usually solved with autoregressive sequence-to-sequence models. However, approaches of this class are inherently slow due to one-by-one token generation, so non-autoregressive alternatives are needed. In this work, we propose a novel non-autoregressive approach to GEC that decouples the architecture into a permutation network that outputs a self-attention weight matrix that can be used in beam search to find the best permutation of input tokens (with auxiliary \textlessins\textgreater tokens) and a decoder network based on a step-unrolled denoising autoencoder that fills in specific tokens. This allows us to find the token permutation after only one forward pass of the permutation network, avoiding autoregressive constructions. We show that the resulting network improves over previously known non-autoregressive methods for GEC and reaches the level of autoregressive methods that do not use language-specific synthetic data generation methods. Our results are supported by a comprehensive experimental validation on the ConLL-2014 and BEA datasets and an extensive ablation study that supports our architectural and algorithmic choices.
%R 10.18653/v1/2023.acl-long.86
%U https://aclanthology.org/2023.acl-long.86
%U https://doi.org/10.18653/v1/2023.acl-long.86
%P 1546-1558
Markdown (Informal)
[GEC-DePenD: Non-Autoregressive Grammatical Error Correction with Decoupled Permutation and Decoding](https://aclanthology.org/2023.acl-long.86) (Yakovlev et al., ACL 2023)
ACL